The field of the invention is medicine agents and methods for combatting atherogenesis.
The role of cholesterol oxidation products in atherogenesis has long been a controversial topic. Peng and Taylor (1983), for example, hypothesized that cholesterol oxidation products may be responsible for an initial arterial cell injury that eventually results in atherosclerosis. On the other hand, Higley et al. (1986) disputed that hypothesis, claiming instead that oxidized cholesterol is substantially less atherogenic than purified cholesterol. The analysis of the oxidized cholesterol shown in Table 1 of Higley et al. indicated the presence of diols and epoxides (78%) as the major oxidized compounds with a minor amount of hydroperoxide (18%) identified as 7xcex1-hydroperoxide.
Hypercholesteroloemia is widely considered to be a major risk factor for the development of atherosclerosis. The lowering of blood cholesterol levels has therefore been an important goal in the search for ways to prevent or treat atherosclerosis. Medicinal agents reported as reducing development of athersclerotic lesions typically result in the lowering of blood cholesterol levels. Beneficial agents which do not affect blood cholesterol are rare. But Bell and Schaub (1986) have reported that chlorpromazine reduced the development of such lesions in rabbits fed an atherogenic diet without lowering of blood cholesterol. Chlorpromazine may function as an inhibitor of calmodulin (Gietzen, 1986) but it is classified pharmacologically as a tranquilizer and sedative (Merck Index, 1976), and therefore is not likely to be useful as a treatment for atherosclerosis.
Tipton, et al. (1987), have reported that a mixture of cholesterol auto-oxidation products prepared from an aged sample of cholesterol inhibits calmodulin irreversibly in a Ca2+-dependent reaction. The reactive material was destroyed by chemical reduction. It was concluded that the calmodulin inhibition was due to one or more cholesterol hydroperoxides. The mixture of cholesterol hydroperoxides was further purified by chromatography and used in feeding experiments with rabbits. Three diets were compared, one with added cholesterol, one with added cholesterol hydroperoxides, and the third with both the cholesterol and cholesterol hydroperoxides. The diet containing cholesterol alone was found to have caused extensive atheroma formation, while the addition of the cholesterol hydroperoxides to the cholesterol diet markedly reduced atheroma formation. The cholesterol hydroperoxides were not found to lower cholesterol concentration in blood plasma, liver or heart. In a related experiment, it was found that the chemically reduced hydroperoxides were not effective in reducing atheroma formation.
Cholesterol oxidizes readily in contact with air, and the oxidation proceeds at ambient room temperature. The oxidation products as initially formed are largely cholesterol hydroperoxides. See Smith, xe2x80x9cCholesterol Autooxidationxe2x80x9d (1981, Plenum Press, New York). This reference lists the initial auto-oxidation products in Table 11, pages 238 to 239. These included 3xcex2-hydroxycholest-5-ene-7xcex1-hydroperoxide; 3xcex2-hydroxycholest-5-ene-7xcex2-hydroperoxide; and 3xcex2-hydroxycholest-5-ene-25-hydroperoxide, which is presently preferred for use in this invention. Other hydroperoxides included 3xcex2-hydroxycholest-5-ene-17-hydroperoxides; 3xcex2-hydroxycholest-5-ene-20-hydroperoxides; 3xcex2-hydroxycholest-5-ene-22-hydroperoxides; and 3xcex2-hydroxycholest-5-ene-24-hydroperoxides.
When it is desired to accelerate the oxidation of cholesterol to hydroperoxides, a photochemical oxidation procedure can be employed as described by Schenk et al. (1958). This reference describes the preparation and isolation of 3xcex2-hydroxycholest-5-ene-7xcex1-hydroperoxide. The 7-xcex2 isomer of this compound can be synthesized and isolated as described by Teng et al. (1973). Other cholesterol hydroperoxides can be prepared as an oxidized mixture, and separated by fractionation. See van Lier et al. (1970).
This invention is based in part on the discovery of a novel class of cholesterol hydroperoxide compounds in which the hydroperoxide group is in the 20-position and the compounds also contain other added oxygenation. As far as is known, no oxidized derivatives of cholesterol have heretofore been reported containing a 20-hydroperoxy group together with a 25-hydroxyl group. An example of such novel compounds is 20-hydroperoxy-25-hydroxycholesterol which has two isomeric forms (R and S). Specific compounds are 20(R)-hydroperoxy-25-hydroxycholesterol and 20(S)-hydroperoxy-25-hydrocholesterol. These compounds can also be named as: 20-(R)-hydroperoxy-cholest-5-ene-3xcex2,25-diol, and 20(S)-hydroperoxy-cholest-5-ene-3xcex2,25-diol.  hydroperoxy groups at two positions. An example of such novel compounds is 20,25-dihydroperoxy cholesterol which has two isomeric forms (20 R and 20 S). Specific compounds are 20-(R)-20,25-dihydroperoxycholesterol and 20-(S)-20,25-dihydroperoxycholesterol. These compounds can also be named as: 20-(R)-20,25-dihydroperoxy-3xcex2-hydrocholest-5-ene and 20(S)-20,25-dihyroperoxy-3xcex2-hydroxycholest-5-ene.
The 20-hydroperoxy cholesterol compounds of this invention are highly active as calmodulin inhibitors, and also display a high level of activity as inhibitors of acyl-CoA:cholesterol acyl transferase (ACAT). ACAT is an index for anticholesterol therapeutic agents (Suckling et al., 1985). The novel compounds of this invention are therefore believed to be useful as therapeutic agents for suppressing atherogenesis in human patients.
The starting material is cholesterol that has been xe2x80x9cagedxe2x80x9d in air at room temperature for a long time (in this case 20 years) or cholesterol that has been heated at 55-60 C. in the dark in a thin layer for 14 days.
The starting material is dissolved in boiling methanol, 30 ml per gram of starting material; the solution is cooled slowly to 4 C. to allow crystallization of unchanged cholesterol. The cholesterol is removed by vacuum filtration and the filtrate is reduced to one eighth the original volume by distillation on a rotary evaporator under reduced pressure. After cooling the concentrated filtrate to 4 C., a second crop of cholesterol is removed by vacuum filtration. The filtrate from this step is extracted three times with hexane and the methanol-rich (lower) layer is evaporated to dryness under vacuum with a rotary evaporator to yield 2.5 g of residue from 50 g of starting material.
This mixture is fractionated by flash chromatography on silicic acid (60-230 mesh, 200 g) eluted with hexane, 2-propanol mixtures according to the following protocol:
Elution was at the rate of 50 mL/min and 50 mL fractions were collected. The desired product elutes in fractions 16-21, with hexane:2-propanol 10:1 v/v. The combined fractions 16-21 are evaporated to dryness to yield about 0.3 g of material which is then rechromatographed on silicic acid (200-400 mesh, 75 g) with hexane and 2-propanol mixtures according to the following protocol:
The flow rate is 20-25 mL/min. After the first 100 mL of solvent, one-minute fractions are collected. Elution of the hydroperoxides is monitored by thin-layer chromatography on silica gel, with benzene and ethyl acetate 3:2 (v.v) as the developing solvent. Hydroperoxides are detected by spraying the plates with 1% N,N-dimethyl-p-phenylenediamine in methanol and water 1:1 (v/v) containing 1% acetic acid. The desired product elutes in fractions 25-31, with hexane:2-propanol 8:1 v/v and the yield is 40 mg.
This mixture is then fractionated by reverse-phase high-performance liquid chromatography on an Ultrasphere (Beckman Instrument Co.) octadecylsilane-treated silica column, 4.6 mmxc3x9725 cm. Elution is with acetonitrile:water 6:4 v/v and the elution is monitored by measuring absorbance at 210 nm. Major peaks eluting at about 12.6 and 15.8 min are collected and designated compounds A and B, respectively. The yields from 50 g xe2x80x9cagedxe2x80x9d cholesterol are 8.9 mg A and 11.9 mg B.
The inhibition of calmodulin activity by A and B were measured using the method described by Leung et al. (1984) and the results are shown in Table 1.
Mass spectra of compounds A and B were obtained by desorption/chemical ionization in isobutane. The fragments used in elucidating the structure are listed in Table 2.
The two compounds have very similar fragmentation patterns, suggesting that they are isomers. The fragments listed suggest that the molecular weight is 434, produced by the addition of three atoms of oxygen to cholesterol. Fragments with m/z 417, 399, 401, 383, and 365 are produced by losses of one or two molecules of water, and/or loss of a molecule of hydrogen peroxide, from the (M+1) ion, m/z 435. The fragment with m/z 317 is produced by Oxe2x80x94O cleavage of the xe2x80x94OOH group at C-20 combined with cleavage of the C-20, C-22 bond; m/z 299 is formed by loss of water from this ion. The fragment with m/z 273 is formed by loss of the side-chain (cleavage between C-17 and C-20) and m/z 255 is formed by loss of water from m/z 273. These fragments show that only the side-chain has been modified in formation of these compounds from cholesterol, that three oxygen atoms have been added, in the form of an alcohol and a hydroperoxy group, and that the most likely site of attachment of the hydroperoxy group is at C-20.  are best interpreted as arising from fragmentation of a molecule of molecular weight 450, produced by the addition of four atoms of oxygen to cholesterol m/z 433 (M+1xe2x80x94H2O); 417(M+1xe2x80x94H2O2); 399(M+1xe2x80x94H2Oxe2x80x94H2O2); 365(M+1xe2x80x94H2Oxe2x80x942H2O2; 317(M+1xe2x80x94OHxe2x80x94C22 to C27; 299(317xe2x80x94H2O); 273(Mxe2x80x94C20 to C27); 255(273xe2x80x94H2O). These fragments show that only the side-chain has been modified in formation of these compounds from cholesterol, and that four atoms of oxygen have been added, in the form of two hydroperoxy groups.
1H-NMR spectra of the compounds and of cholesterol were obtained and are compared with the spectra of two diastereomers of 20-hydroperoxycholesterol, reported by van Lier and Rousseau (19760, in Table 3.
The signal for C-21, which is a doublet in the spectrum of cholesterol, is a singlet in the spectra of compounds A and B and the isomers of 20-hydroperoxycholesterol, showing that the proton on C-20 has been replaced. Similarly, the signals for C-26 and C-27, doublets in the spectra of cholesterol and 20-hydroperoxycholesterol, coalesce to a singlet in the spectra of compounds A and B, showing that the proton on C-25 has been replaced.
The hydroperoxy group is shown to be at C-20, and the hydroxy group at C-25, by the following comparisons (Tables 3 and 4).  The stereochemistry at C-20 is deduced by the following comparisons (Tables 3 and 4).
The 1H-NMR spectra of 20-xcex1- and xcex2-hydroxycholesterols have been compared (Honda and Komori, 1986; Nes and Varkey, 1976). Both epimers show the same chemical shift for protons on C-18, while the protons on C-21 have different chemical shifts in the two epimers. On the other hand, van Lier and Rousseau report that in the two epimers of 20-hydroperoxycholesterol, the chemical shift for the C-21 protons is the same, while the chemical shift for the C-18 protons is affected by the stereochemistry at C-20. Compounds A and B have the same chemical shift at C-21 but differ at C-18 while the products of reduction of the hydroperoxides to alcohols have the same chemical shift at C-18 and differ at C-21. Therefore the hydroperoxy group is at C-20. The stereochemistry at C-20 of Compound A and B can also be deduced from the data in Tables 3 and 4.  The chemical shift at C-18 of compound A is further upfield than that of compound B and therefore by analogy with the epimers of 20-hydroperoxycholesterol, compound A is the xcex1 isomer (cholest-5-ene-3xcex2,20S-diol)  xcex1 isomer (20-(S)-20,25-dihdroperoxy-3xcex2-hydroxycholest-5-ene), and compound B is the xcex2 isomer (cholest-5-ene-3xcex2,20R-diol)  xcex2 isomer (20-(R)-20,25-dihydroperoxy-3xcex2-hydroxycholest-5-ene). Comparison of the spectra of the reduction products to those of the epimers of 20-hydroxycholesterol confirms this assignment of stereochemistry.
The novel 20-hydroperoxide cholesterol compounds are prepared, purified and isolated as described above for use as therapeutic agents in suppressing atherogenesis in human patients. The preferred compounds are 20(S)-hydroperoxy-cholest-5-ene-3xcex2,25-diol and 20(R)-hydroperoxy-cholest-5-ene-3xcex2-25 diol.  The preferred compounds are 20-(S)-20,25-dihydroperoxy-3xcex2-hydroxycholest-5-ene and 20-(R)-20,25-dihydroperoxy-3xcex2-hydroxycholest-5-ene. These compounds may be used separately or in admixture. 20-hydroperoxy cholesterol compounds of this invention are characterized by being potent inhibitors of calmodulin, for example, by the assay described by Leung et al. (1984). They also demonstrate a high level of inhibitory activity in the ACAT assay, as described by Mather, et al. (1985).
The amount of cholesterol 20-hydroperoxide to be administered can be related to the body weight and/or the blood cholesterol level of the patient. A sufficient amount of the cholesterol hydroperoxide should be administered to be effective for reducing aortal deposit of cholesterol. For example, from 0.5 to 25 milligrams (mg) of cholesterol 20-hydroperoxide can be orally administered per kilogram (kg) of body weight per 24 hours. The administration should be on a regular basis for a period of time as required to prevent or treat cholesterol-induced atherogenesis. The cholesterol 20-hydroperoxide may be prepared in the form of tablets or capsules. The 20-hydroperoxide compounds can be combined with tabletting compositions, viz. dextrose or sucrose, and formed into tablets. Conveniently, the tablets may contain an amount of the cholesterol hydroperoxide so that 2 to 4 tablets per 24 hours may be taken. For example, each tablet may contain 12.5 mg for a two tablet dose, or 6.25 mg of cholesterol hydroperoxide for a four tablet dose per 24 hours, or other effective dose amounts.
The cholesterol-20-hydroperoxides may also be prepared in capsule dose form. The capsules can contain only the cholesterol 20-hydroperoxide, or it may be admixed with a pharmaceutical extender or diluent, such as lactose, dextrose, etc.
For oral administration, either in the form of tablets or capsules, the hydroperoxides may be administered in doses of from about 0.5 to 25 mg of cholesterol 20-hydroperoxide per patient per 24 hours. If the cholesterol 20-hydroperoxide composition includes oxidized derivatives of cholesterol other than the 20-hydroperoxides, dosage should be on the basis of the 20-hydroperoxide content only.
The administration of cholesterol 20-hydroperoxide can be monitored if desired. For example, a blood sample can be obtained and the serum separated with the lipoproteins therein. A portion of the serum can then be subjected to a cAMP-phosphodiesterase assay (Sharma and Wang, 1979) to measure calmodulin inhibition.